A recent article over at Do512 reminded me to check for new migration data from the IRS, and, unfortunately, we’re still waiting on 2012. While we wait, here’s a refresher course on the extent to which we can blame California for worsening traffic, rising home prices and rents, or any of the other consequences of Austin’s growth and economic success.
There are many. The two most commonly used sources of data on migration are the U.S. Census Bureau’s American Community Survey (ACS) and tax records from the IRS. ACS estimates are derived from samples and therefore have margins of error to deal with, and IRS data excludes people who don’t file taxes. Both have considerable time lags, and are subject to regular attacks by free-dum loving politicians, jeopardizing future availability. ACS provides summary statistics for how many people move in and out of cities, but not much on specific origins or destinations. IRS provides county-to-county looks, but nothing for cities.
So, the best we can do to get a handle on the impact of migration from California to Austin (city) is to use ACS and IRS data and look at Travis County. Hopefully one day we will find a client to foot the bill for exploring the reliability of migration data from private vendors, or perhaps even collect our own primary data for Austin, but for now we are limited to what’s available from public sources.
This is different from challenged interpreters, who are usually politicians and boosters emphasizing move-ins and neglecting move-outs. Believe it or not, people do actually decide to move away from Texas, and, yes, even Austin. Interpretation challenges include things like boomerang migration–e.g., somebody moves from San Francisco to Austin, toughs out two summers here while developing a newfound appreciation for BART, and then moves back to San Francisco. That person shows up in the data as one “Californian” moving to Travis County and then one “Austinite” moving to California. Or, to complicate even further, make that person an Austin native who moves to San Francisco, can’t find acceptable brisket or $2 Lone Star anywhere, and then moves back to Austin. He shows up in the data as a “Californian” moving to Travis County, when, really, the two moves should cancel each other out, at least in how we commonly perceive and talk about migration’s impact on Austin.
Same goes for what Ryan Robinson, City of Austin’s demographer, calls “two-step” migration. Suppose somebody moves from San Francisco to Dallas and then the following year to Austin. She shows up in the data as a California transplant to Dallas County in year one, but then as a Texas transplant to Travis County in year two. She may even still have California plates, but, according to the data, she’s a Texan.
And there are other limitations to be aware of. For example, IRS migration statistics are reported in tax terms–i.e. number of returns and exemptions as reported on tax filings, not households and people. Most researchers use returns as proxies for households and exemptions as proxies for people, but they are not one-to-one comparisons (e.g., a married couple filing separately can be two returns but one household). Further, the IRS goes to great lengths with statistical procedures employed to maintain confidentiality. Consequently, there are missing values, especially for small and rural counties.
While you may feel perfectly comfortable drawing conclusions about how wealthy your new neighbor from California is by looking at home prices on Zillow or TCAD, being able to derive her household income from IRS migration data would be bad. We can’t track movements of unique households or individuals over time, either, just in aggregate snapshots, which makes migration analysis difficult but keeps the purveyors of freedom in Congress at bay.
The Californicators (I miss John Kelso’s regular column)
How many Californians are invading Austin? Somewhere between approximately 58,000 (IRS) and 81,000 (ACS) people moved to Travis County from another county in the U.S., on average, per year during 2008-2012. Roughly 60% of those new residents in Travis County came from some other county in Texas. The estimated number of people moving from California to Travis County annually between 2008 and 2012 ranged from approximately 3,800 (IRS, 2011 reporting year) to 4,400 (ACS). So that means California was responsible for about 15% of the 24,000 (IRS) to 29,700 (ACS) moving to Travis County from outside Texas on average per year during 2008-12.
California is, by far, the most significant “donor” state of residents to Travis County, but still pales in comparison to the number of people moving here from other parts of Texas. Further, California doesn’t make up anything close to a majority of people moving to Travis County from other states–for every seven people you meet who have moved here from another state, only one of them is likely to be from California.
Now, what happens when we account for people moving out of Travis County? Estimates for people moving out of Travis County to another county in the U.S. ranged from approximately 50,900 (IRS) to 69,000 (ACS), on average, per year during 2008-2012. Travis County usually breaks even, more or less, with the rest of Texas–i.e. the number of people moving in to Travis County from another county in Texas is roughly equivalent to the number of people moving out of Travis County to another county in Texas. That can change in some years depending on local economic conditions, but over the long term and looking at both IRS and ACS data it’s roughly a stalemate, with Travis County gaining at most 1,000-1,500 people per year.
Net migration (move-ins – move-outs) from other states is more of a factor in Travis County’s overall population growth. Travis County gained, on average, somewhere in the range of 8,505 (IRS) to 9,200 (ACS) people per year from other states during 2008-2012, including approximately 1,000 (ACS) to 1,400 (IRS) from California. To put that into perspective, average attendance at an Austin Aztex home game is 2,687. So, using the high end of the range, the number of people Travis County is gaining (net) from California every year is equivalent to about one-half the number of people you will find at an Aztex game. Not exactly what I’d call a population “driver.”
Well can we at least blame them for rising home prices?
Sadly, this, too, is problematic, or at least unclear. Using the IRS migration data (2011), adjusted gross income per return (household) for people moving from California to Travis County was $78,384, well above the $67,118 for existing Travis County residents (non-movers), but well below some other states.
Illinois, for example, sends one-third as many people to Travis County as California, but adjusted gross income per return for Illinois transplants was nearly twice that of California ($148,173), according to IRS data (2011). In fact, people moving to Travis County from seven other states had higher adjusted gross incomes per return than people moving to Travis County from California that year. While the aggregate impact of more Californians with higher than average incomes moving to Travis County may outweigh the impact of other, even higher income states with fewer transplants, Californians are not the wealthiest people moving here, at least on average.
Visit Census Flows Mapper for breakdowns of migration by income, race/ethnicity, educational attainment, employment status, and more. Perhaps you can even shed some light on the reported “mass exodus” of Austin residents.